Australia's Say-on-Pay Rule: Pay-Performance Link and Auditors' Response

Over the last three decades, executive compensation has attracted considerable media attention worldwide and has become a significant issue in corporate governance debates among academic and business communities. Specifically, due to corporate collapses and scandals in the early 2000s as well as the global financial crisis of 2008, managerial remuneration has been subjected to much criticism in recent years. Regulators and activist shareholders around the world have advocated that shareholders have a greater say (vote) on executive pay and greater influence in the boardroom to restrain managerial power and rent extraction. Therefore, Say-on-Pay (SoP) legislation has been enacted in several countries including the United Kingdom, the United States, France, Germany, Japan, Canada, Brazil, China, Russia, India, Belgium, Spain, Sweden, Switzerland, the Netherlands, Finland, Denmark, Norway, Italy, South Africa, Israel, and Slovenia. Australia also has introduced SoP regulation, namely the ‘two-strikes’ rule in 2011 after a decade of growing concerns with the ‘fat cat’ pay and rewards for failure. Australia’s ‘two-strikes’ rule is viewed as a unique and innovative governance tool in that it empowers minority shareholders to remove a board of directors if a listed company receives a ‘strike’ against its remuneration report in two consecutive years. However, giving shareholders more power to have a greater direct say over the executive pay have been heavily debated. Proponents of SoP laws argued that allowing shareholders votes on executive compensation would increase accountability and transparency and incentivize boards to fulfil their fiduciary duty and ensure that CEOs are paid for performance, and thereby maximise shareholder value. In contrast, opponents argued that SoP rules would empower shareholders who lack the required expertise and sophistication, and therefore are likely to abuse their voting rights and put unnecessary pressure on the board of directors to adopt compensation contracts that may potentially destroy firm value. Given that Australia has moved away from the CLERP9 advisory vote to Australia’s ‘two-strikes’ rule, an important policy question to address is whether this change has had a positive or negative effect and whether this transition achieved or failed the intended goal to align the interests of management with those of shareholders by improving the relationship between executive compensation and firm performance. Given that the ‘two-strikes’ rule has specific and predictable consequences for Australian firms, this thesis has three objectives. First, this thesis examines the pay-performance relation before and after the introduction of the ‘two-strikes’ rule on ‘non-strike’ firms. Second, this thesis investigates whether the incidence of receiving a ‘first strike’ improves the pay-performance link in the long term. Third, this thesis investigates whether auditors respond to the incidence of receiving the first ‘strike’ by increasing the audit fees, issuing modified audit opinion, or resigning from an audit engagement. Within executive remuneration, CEO pay draws the most attention from the media, politicians, and the public. CEO pay also captures the philosophy and essence of executive remuneration structure. Hence, this thesis focuses on CEO pay. Using a large sample of Australian Securities Exchange (ASX)-listed firms with 9,513 firm-year observations for the period 2004-2016, this thesis estimates both CEO pay-performance sensitivity (PPS) and pay-performance elasticity (PPE). To avoid possible sample selection bias, this thesis employs a propensity score matching (PSM) design to match ‘first strike’ firms (treatment) with ‘non-strike’ (control) firms. As part of robustness tests and to enhance the credibility of the findings, the thesis adopts an alternative control sample. It employs a Matched-Pair Design (MPD) with a three-way matching strategy: financial year, GICS-based economic-sector classification, and firm size. Multivariate panel data estimation and logistic regression techniques were used for analysing data and drawing inferences. Findings in relation to the first research question suggest that, among the ‘nonstrike’ firms, the pay-performance relation was positive and statistically significant before the enactment of the ‘two-strikes’ rule in 2011. Specifically, CEO wealth of Australian firms increased by $51 for every $1000 increase in the shareholder value. With respect to pay-performance elasticity for Australian firms, CEO remuneration increased by 8% for each 10% increase in shareholder value. However, the relationship was reversed after the introduction of the ‘two-strikes’ rule. Specifically, CEOs of the Australian firms experienced pay cuts of about $12 for each $1,000 increase in shareholder value post-2011. With respect to pay-performance elasticity, CEO pay decreased by 0.78% for each 10% increase in shareholder value during the 2012-2016 period. The results for both sensitivity and elasticity support the interference hypothesis that the implementation of the ‘two-strikes’ rule with the threat of removing the boards in a very public way may have put unnecessary pressure on Australian boards to punish their CEOs more severely than they deserved. The reversal of the pay-performance link post-2011 can be explained as the following. There is no collective definitive view of the shareholders about the remuneration report until the annual general meeting takes place. A ‘first strike’ itself can bring a firm under negative spotlight. Further, it can be viewed as a reprimand of the board’s poor corporate governance practice. A ‘first strike’ can suggest the presence of high agency costs in the firm signalling other potential problems to shareholders, lenders, and auditors. Regulators might also take interest in the firm to have a closer scrutiny of its governance practice. All things considered; most firms listed on the ASX might have taken a pre-emptive measure of avoiding a ‘first strike’ by significantly reducing CEO compensation while improving firm performance. This could create a negative relation between CEO pay and firm performance in the ASX-listed firms after the implementation of the ‘two-strikes’ rule (2012-2016). Findings regarding the second research question suggest that, unlike the ‘nonstrike’ firms, the estimated sensitivity of CEO compensation of the ‘first strike’ firms with respect to shareholder wealth following the ‘first strike’ were all negative and significant. The sum of the shareholder wealth change coefficients implies that the wealth of CEOs of the ‘first strike’ firms decreased by about $15 whenever shareholder wealth increased by $1,000 during 2012-2016. With respect to the elasticity model, the results indicate that for a 10% increase in shareholder wealth, the CEO received pay cuts of about 0.12%. By examining the significance of coefficients on the interactions between the ‘first strike’ and changes in shareholder wealth, this thesis finds evidence that receiving a ‘first strike’ is not significantly associated with pay-performance sensitivity or elasticity in the year of the ‘first strike’. Overall, the results suggest that Australian shareholders may focus more on how much executives are paid (the total of CEO remuneration) rather than how they are paid (the pay-performance link). Furthermore, the results imply that after receiving the ‘first strike’ boards may punish CEOs more severely than they deserve by reducing future remuneration even though shareholder wealth is increasing in the long run. Thus, allowing minority shareholders to remove the boards may put public pressure on Australian firms and lead to the unintended consequence of decreasing the total CEO compensation, rather than improving the alignment of executive pay and firm performance in the long run. Further, the results suggest that shareholders may not have the sophistication necessary to use their new voting power and to target firms with weaker pay-performance relation. Thus, it might be argued that the dissatisfaction of minority shareholders may not reflect specific concerns about poor remuneration design but rather other issues such as capital raising, declining share price, and poor or unexpected financial results. Findings with regard to the third research question suggest that relative to ‘nonstrike’ firms, auditors were more likely to increase audit fees, resign or be removed from the office in the year following the ‘first strike’. Furthermore, the thesis finds that firms receiving a ‘first strike’ tend to have higher incidence of modified audit opinions than control firms. This result suggests that the incidence of a ‘first strike’ may signal clients’ higher business and/or audit risk to the auditors. In sum, the findings support the notion that the public scrutiny associated with campaigns by shareholder activists heightens auditors’ concerns about reputational damage and litigation risk. The results also suggest that the external auditors should use the incidence of a ‘first strike’ as an input in their risk assessments. This thesis contributes to the CEO compensation literature in general and the pay-performance link literature in particular by investigating the pay-performance link in preand post-implementation periods of Australia’s ‘two-strikes’ rule (2005-2011 and 2012-2016, respectively). The findings of the thesis have important policy implications for other countries that are in the process of adopting some form of SoP regulation. The findings are important in informing investors, analysts, and managers that shareholder activism in the form of votes on the remuneration report can have diverse consequences. Moreover, awarding ‘excessive’ power to minority shareholders can lead to unintended consequences. The findings of this thesis would have important implications for boards in terms of taking defensive actions to avoid a ‘first strike’ against r

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.